The Effect of Twitter Attention on Earnings Announcements. An Event Study Investigating the Relation Between Twitter Attention, by Volume and Sentiment, and Market Efficiency


Publication date



Oslo and Akershus University College of Applied Sciences

Document type


Master i økonomi og administrasjon


In our thesis, we use event study to analyze relations between Twitter attention and technology companies on the Nasdaq stock exchange. We divide attention into Twitter sentiment and volume. The classification of sentiment is done with supervised machine learning algorithms. The effects of Twitter sentiment are consistent with the Efficient Market Hypothesis. Twitter volume on the other hand, indicates an attention effect related to abnormal market behavior, mainly in larger well-known companies. We believe an indicator as to how individual investors pick stocks, is to distinguish whether a firm’s operations are known to the investors, and how abnormal tweet volume affect traders’ attention towards them. Handelshøyskolen ved HiOA Oslo




Permanent URL (for citation purposes)